Quantifying and predicting air quality on different road types in urban environments using mobile monitoring and automated machine learning

Abstract

Traffic emissions are a primary source of air pollution in urban areas, with air quality being influenced by different types of roads characterized by varying traffic volumes and speeds. Comprehending the distribution of air pollutants and the factors influencing it across different road types holds immense significance in endeavors to enhance air quality within urbanized regions. This study recorded concentrations of PM, SO2, NO2, CO, and O3 on different road types in Shenyang, China, using mobile monitoring. The impacts of road type and microclimatic factors on air quality were quantified using automated machine learning. Among the six road types, the suburban highway exhibited the highest PM, SO2, and NO2 pollution. On the other hand, secondary roads experienced the highest levels of CO and O3 pollution. The automated machine learning models provided accurate predictions for PM2.5, PM10, SO2, NO2, and O3 concentrations (R2 = 0.91, 0.83, 0.82, 0.83, 0.79, respectively). Relative humidity played the most significant role in PM2.5 and PM10 concentrations (55.93% and 59.39%, respectively), followed by air temperature (15.36% and 17.73%) and road types (14.28% and 8.74%). Road types contributed 24.33%, 20.60%, 16.61%, and 11.90% to SO2, CO, O3, and NO2 concentrations, respectively. Overall, this study addresses the limitations of previous research and provides a comprehensive understanding of the impact of road types on air pollutant concentrations in urban environments.